Vol.229 Macro Talk 109 | We Are at the Tail End of an Upward Dollar-Dominated Capital Cycle (recorded Aug. 6)

Summary

• Li Feng’s top-line call: dollar-denominated assets are at the tail end of the upswing in the capital cycle—with no large-scale incremental inflows, the market has become a zero-sum rotation. The result is the recent pattern of “limit-up one day, limit-down the next”: money is both afraid to chase highs and eager to chase hot spots, jumping “from one hole to another.” Even within semiconductors, the trade flips from CPU to memory to GPGPU. This does not mean an immediate selloff, he stressed, but “the last wave of money is very hard to make.” In a broad market decline, almost no narrow category can keep rising independently for long.
• The US fiscal ledger has squeezed the country into a passage “thinner than a tightrope”: rate hikes raise interest costs, falling inflation offers little room for cuts, and the Treasury still has to issue debt aggressively. US debt is about to break $40T; July’s annualized interest bill was roughly $1.45T, and adding around $1.4T in defense spending brings the total close to $3T—more than half of federal revenue of over $5T. The 10-year yield is around 4.7% and the 30-year has broken above 5.2%. Issuance is especially heavy from June through September, with roughly $80B mentioned; later, the discussion says more than $3T still needs to be issued over the next 3 months. Central banks, led by China, are selling Treasuries and adding gold.
• Three visible black swans are emerging: a reversal in yen liquidity, private debt financing for data centers, and the Strait of Hormuz. A day and a half of Japanese FX intervention could dump $50B-$60B, after more than $100B has already been spent this year; if Japan is forced into rapid rate hikes, carry trades could unwind and Japanese capital could return home, creating “a bigger black swan for other types of risk assets.” America’s most profitable companies are shifting from hidden buyers of Treasuries to issuers scrambling for liquidity: Google’s cash flow turned negative in Q2, as did that of most large internet companies except Microsoft, while Oracle’s private-debt rates keep rising. If Iran’s control of Hormuz halts shipping for another 70 days, commodities could collapse. The negative correlation between Treasuries and US stocks is near a historical extreme: “Black swans never start with earnings.”
• The cycle framework is straightforward: broad-based gains across asset classes require ever-larger net inflows, but the world has “not much money left to siphon off.” US equities rose from just over $30T in 2019 to roughly $70T; Li Feng estimates that this required $5T-$8T of inflows. US debt rose from $21T to $40T, while global debt is moving from $250T toward more than $350T. The $15T-$20T of base money released in 2020-2021 was expanded 3-4x through the money multiplier, driving the 2023-2025 migration into dollar assets. The other side of a weaker dollar is gold.
• China’s domestic focus is on 3 points from July’s Politburo economic work meeting: services consumption, social-security coverage for flexible workers, and “facing the difficulties at hand,” while the property language shifted to the unqualified phrase “stabilize the property market.” After a Qiushi article called for repairing household balance sheets, policies implemented on Aug. 1 were widely interpreted as restricting multiple internet loans—a one-time cleanup of the liability side, “like stripping all the bad loans out of the Big Four banks in one shot in 2000.” How the RMB100B of fiscal funding will be leveraged is a matter of speculation, with interest subsidies the leading theory. One possible target is first-home mortgages: a 1-percentage-point subsidy could push mortgage rates below the average rent-to-price ratio in Tier 1 and Tier 2 cities, and below 2%.
• Micro evidence of a property stabilization is emerging first in the oldest, most run-down homes in Tier 1 and Tier 2 urban districts, where the rent-to-price ratio has reached roughly 2.6 and has been reported as above term-deposit rates, potentially drawing investment demand back slowly and temporarily. The long-run steady state is for home prices to rise slightly faster than inflation: with CPI at 1%, annual gains of roughly 1.3%-1.5% remain consumption-driven, while investment demand begins to increase above 2.3%. Li Xiang asked why first-home mortgage rates could not simply be cut to zero. Li Feng’s answer: bank net interest margins are already at historical lows, and zero rates would mean relying “entirely on fiscal policy”—possibly effective in the short term but difficult to sustain.
• China’s assets are unusually difficult to forecast in one direction because stabilization operations, savings activation, and capital controls provide a buffer. Huijin and China Securities Finance had sold more than 90% of their positions before early July, “bringing all the ammunition back”; they then bought broad-market ETFs aggressively on oversold days and sold them back once the market stabilized intraday, suggesting algorithmic or trigger-based execution. Foreign investors remain underweight Hong Kong equities. If excess savings—estimated at anywhere from RMB10T-plus to more than RMB30T—are activated, the money will create large, sustained inflows wherever it goes. Primary-market heat remains highly concentrated in embodied intelligence, world models, quantum computing, and controllable nuclear fusion. If the US capital cycle tops out and the market plunges, investment could instead spread across genuinely monetizable applications of “AI+” in consumer, hardware, industrial, and manufacturing sectors. For Hong Kong stocks, his only personal gauge is the rate of first-day and seven-day breaks below issue price: only when that rate rises can the market “become more rational and more investment-oriented.”

Deep dive

1. A Checkup on the Capital Cycle’s Final Stretch: No Incremental Money, Just Wide-Ranging Rotation

• Li Xiang opened with an observation about the frequency and intensity of volatility: “limit-up one day, limit-down the next.” Li Feng’s framework is that sustained gains require large-scale positive incremental inflows. Once the total pool stops growing, the final stage brings “fear of heights” and hot-spot chasing at the same time—the same money jumps “from one hole to another.”
• The tell is that indexes no longer rise together. Instead, A goes up while B falls, with both the speed and magnitude of sector rotation increasing. Even within semiconductors, the money moves from CPU to memory to GPGPU: speculative capital is running from trade to trade, while no fresh money is coming in to lift the market as a whole.

2. July’s Politburo Economic Work Meeting: Several Phrases Drew Attention

• The backdrop was a Q2 increase in fiscal spending that lagged revenue growth by a wide margin, along with government investment running behind schedule. Local governments were waiting for signals on the second half. The communiqué drew attention for elevating services consumption, giving social security for flexible workers its own heading, and explicitly calling for “facing the difficulties at hand”—a direct acknowledgment of this year’s economic problems.
• The technology language was in line with expectations, but the emphasis was on “AI+,” meaning applications. In property, the wording became the unqualified “stabilize the property market.” “That sentence is already a conclusion,” Li Feng said; it no longer says that policymakers will “work to stabilize property-market expectations.”

3. The Overcapacity Debate and Trade Structure

• Li Feng’s counterargument is that if China’s capacity were excessive for the world as a whole, every economy should be experiencing mild deflation or mild inflation—in other words, low inflation or limited deflation. The reality is that almost every country outside China, developed and emerging alike, is dealing with inflation. “At a minimum, you cannot say Chinese capacity is excessive for the world.”
• The meeting’s call to “balance the trade structure” means shifting from low- and mid-value-added goods toward mid- and high-value-added goods. It also encompasses a previously discussed path: keep the RMB stable while allowing modest appreciation against currencies other than the dollar, thereby increasing imports.

4. Services Consumption, Flexible Work, and the Three-Way Intersection with Urbanization

• Services consumption is labor-intensive, creates more jobs, and raises the share of labor compensation in GDP or corporate sales. But China’s services expansion is arriving alongside digitalization, platformization, and eventually AI. Employment therefore looks less like the individual dentists and small law practices that emerged in the US and more like “professional service providers on platforms.” Most services workers live in cities, directly connecting the issue to the urban integration and social-security needs of new residents.
• Platformization partly addresses the Baumol effect—the displacement created when labor moves from high-productivity manufacturing into lower-productivity services—but creates “a small socioeconomic problem”: does a platform count as the employer of a flexible worker? One person may deliver food for platform A, parcels for platform B, and work in a restaurant, with no clear answer on who should pay what share of social security. Food-delivery platforms began piloting cost-sharing with workers after the pandemic, but multiple platforms and mixed income streams—tips, livestream-commerce revenue shares, and employment by MCNs—remain unresolved. “The US clearly has no better solution.”
• A newer model is the OPC, or one-person company, with AI agents running accounts and generating income across multiple platforms. “Looked at from the left, it eases some employment pressure; looked at from the right, it is also a new form of flexible work.”

5. Repairing Household Balance Sheets—The Liability-Side Cleanup on Aug. 1

• After a Qiushi article called for repairing household balance sheets, many interpreted the policies implemented on Aug. 1 as restrictions on multiple internet loans. That came on top of the lending-rate crackdown that began last October: annualized rates must stay within 24%, and rates above 4x LPR receive no legal protection. Li Feng said internet-finance companies, including assisted-lending platforms, may have fallen 60%-70% or more since April or May. The remaining P2P businesses tied to lending—fixed-income and credit-claim products—also appear to have undergone a final cleanup.
• Li Feng’s analogy was blunt: “This is like the banking reform of 2000, when we stripped all the bad loans out of the Big Four banks in one shot.” Financial regulation is being used to clear out high-interest borrowing by the weakest borrowers “in a very short period.” He expects bank retail NPL ratios to swing sharply in the short term, while banks are also being tightly restricted from taking “referral” loans from third-party platforms.

6. The Mystery of Falling Leverage: RMB50T of Deposits Maturing and Early Loan Repayment

• Li Feng called it “very strange”: continued property depreciation should have pushed up household leverage as the denominator—assets—shrank, yet household leverage fell in the first half because liabilities declined faster than assets. That fits his earlier view that roughly RMB50T of household deposits, mainly time deposits, will mature this year. With no high-yield products available, many households are using the money to repay short- and long-term commercial loans whose rates are higher, explaining why household long-term borrowing fell rather than rose even after property prices rallied in March.

7. Why “Stabilize the Property Market” Matters for Consumption: The Mental Account

• Repairing household balance sheets means shrinking the liability numerator while preventing the asset denominator from shrinking at the same time. The goal is to expand domestic demand: “When you feel your wealth declining and shrinking, your consumption is naturally conservative.” Someone who bought a home for RMB3M, watched it rise to RMB12M, and then saw it fall back to RMB8M feels that he has lost RMB4M—the same mental accounting that treats a prior RMB10M stock-market gain as principal.
• Li Xiang’s field observations were consistent: he spoke with 2 or 3 owners of small companies with a dozen or 20 employees whose revenue and profit had not fallen but who were cutting spending. “It feels as if I have to use every ounce of effort to achieve today’s revenue and profit, so we lower our expectations.” Li Feng said this is a matter of psychological expectations, shaped by the broader environment and surrounding information.

8. Old, Run-Down Homes Stabilize First: A 2.6 Rent-to-Price Ratio Signal

• A Sanlian Life Week survey covering roughly 10 Tier 1 and Tier 2 cities, including Shenzhen, Wuhan, Chongqing, Shanghai, and Hangzhou, found that the old, run-down homes in urban districts—those hit hardest over the past 3 years—stabilized first after 2.5 rounds of declines. Their rent-to-price ratio reached roughly 2.6, which reports said exceeded the 3.7% term-deposit rate and even the interest income on longer-dated deposits. Both prices and transaction volumes stabilized. Li Feng guessed that cities such as Shanghai “should be starting to see some investment demand.”
• A marginal factor was the purchase of rental housing by local governments in some major cities starting in May. Once constraints on floor area, location, and total price are applied, the homes are “almost exactly old, run-down units.” But given the state of local-government finances, the program will not be large and is “only a very small contributing factor.” The main driver is still the rent-to-price ratio. Market capital obeys the same rule: “Whenever an investment product yielding more than 2% appears, it attracts some money.”

9. How Can RMB100B of Fiscal Funding Create Leverage?

• Finance Minister Lan said roughly 2 weeks earlier that RMB100B of this year’s fiscal funds should “produce a better result rather than be paid out as a direct subsidy.” Li Feng’s translation: it needs to create leverage, which conventionally means targeted interest subsidies—“adding targeted interest subsidies on top of commercial rates that are already low.” Market speculation has centered on mobilizing RMB500B or RMB800B. Potential targets include technology innovation, small businesses, and services, with first-home, essential housing also mentioned. A 1-percentage-point subsidy, combined with the provident-fund system, could push mortgage rates below the average rent-to-price ratio in Tier 1 and Tier 2 cities and below 2%, accelerating a bottoming and stabilization.

10. Li Xiang’s Question: Why Not Cut First-Home Mortgage Rates to Zero?

• Li Feng’s answer was that China’s banking-sector net interest margin is at a historical low, and the spread is the banks’ only way to make money. Low rates serve debt resolution on one side and economic stimulus on the other. If lending rates went to zero, even taking deposit rates to zero would leave no spread: it would rely “entirely on interest subsidies, entirely on fiscal policy.” That would amount to deploying a large volume of government debt into one area—possibly effective in the short term, but difficult to sustain. The economic path is to release demand at a reasonable pace and return to a reasonable rent-to-price ratio; subsidies should be an accelerator, not the engine.
• Banks face 3 simultaneous transformation pressures: absorb a buildup of household-loan bad debt in the short term, operate for a period with historically low spreads, and shift some activity toward direct financing in line with the direction set by the 2014 financial reforms and the 2018 asset-management rules. “If the banking sector were a K, it has already moved part of itself onto the upper branch.”

11. Household Finance and the State’s Long-Cycle Transformation Move in Parallel

• China’s biggest macro-financial structural change is a shift from bank loans as the dominant form of financing to a combination of loans and direct financing. It is “definitely, continuously, and over the long term” taking place. Households are moving in the same direction: away from deposits and property, structures linked to indirect financing, and toward insurance, wealth management, stocks, and funds. Roughly 70% of the underlying assets of wealth-management and insurance products are still bonds, and bonds are direct financing. From 2000 through roughly 2016, the state and individuals adjusted in parallel; since 2019 or 2020, they have been “slowly and over the long term” adjusting in the other direction. This is another long cycle.
• Li Xiang recalled an earlier debate over whether the 14th Five-Year Plan should classify property as a consumer good. Li Feng said it has “more of a consumption attribute.” His steady-state rule is that annual price gains slightly above inflation are still owner-occupier driven: with CPI at 1%, roughly 1.3%-1.5% annual appreciation. Investment demand starts to rise above 2.3%. One small signal came in August, when major banks restored quotas for 5-year large-denomination time deposits, whose rates can reach 2% when high. Li Feng guessed that savings activation accelerated in July, putting pressure on banks to attract deposits; July data may show household-deposit growth below last year’s level and a rise in non-bank deposits.

12. The US Ledger: Interest and Defense Consume Half of Fiscal Revenue

• US debt is about to break $40T. The Fed policy rate is unchanged at roughly 3.5%-3.75%, short-term yields are around 4%, the 10-year is about 4.7%, and the 30-year has reached roughly 5.2% or higher—near historical extremes. At a July annualized interest bill of around $1.45T, adding approximately $1.4T in defense spending brings the total close to $3T, more than half of federal revenue of over $5T. “Everything left—whether infrastructure, healthcare, education, or anything else—is being severely crowded out.”
• The problem is compounded by more frequent refinancing and a widening deficit. Issuance is particularly heavy from June through September, with roughly $800B mentioned; in the later discussion of the next 3 months, the figure is more than $3T still to be issued. Central banks around the world, led by China, have been selling Treasuries and buying gold over the past 2 years. “There are already not many buyers in the market.” Warsh has also discussed shrinking the balance sheet and returning the Treasuries held by the Fed to the market.

13. The Yen Deadlock: “A Very Difficult Vicious Circle to Break”

• Japan is one of the major buyers of US Treasuries, but the yen had fallen to a 40-year low. A day and a half of intervention may have consumed $50B-$60B, with more than $100B spent since the start of the year. If Japan intervenes by selling dollars and buying yen rather than hiking rates, it can eventually “only sell Treasuries.” Treasury sales push yields higher, raising US funding costs and closing the loop.
• Three pools of money could flow out of Japan: the enormous overseas assets of Japanese companies and pension funds, US Treasuries held by the Japanese government and BOJ, and carry trades built on near-zero rates in a market without FX or capital controls. A rapid Japanese rate-hike cycle would force carry trades to unwind and sell dollar assets, while Japanese capital would return home to benefit from yen appreciation. “You cannot let Japan fall this far and sell Treasuries, but you also do not want it to hike rates rapidly”—a very difficult vicious circle. The US “claims” to be helping stabilize the yen, but the Treasury can deploy only a few tens of billions of dollars, without congressional debate, to stabilize other countries’ or dollar-related exchange rates. In essence, it is managing expectations. The path of “the dollar depreciating only slightly against the yen” is “extremely, extremely narrow.”

14. The Great Migration of 2023-2025 and the Mini-Cycle of Q2 2026

• The bigger picture Li Feng outlined at the start of the year is that China’s challenges, combined with Europe’s problems following the Russia-Ukraine war, caused the huge volume of money created in excess in 2020-2021 to “move slowly into dollar assets.” It first tested 5%, then rose to 15%, 20%, and 25%, producing the catch-up rally in dollar assets from 2023 through 2025. The current volatility is the tail end of that allocation cycle.
• The mini-cycle is replaying the larger one. After the US-Iran war at the end of February, March brought panic and a move into cash, followed by allocations to highly liquid, high-safety assets. The sharp rise in the dollar index in April and May coincided with a surge in US memory-chip stocks. “The US created a crisis and siphoned the world’s money once again.” That partly explains why China began controlling FX outflows in May. Once the system can no longer siphon money, only wide-ranging volatility remains.

15. Warsh’s Expectations Game: Rate Hikes Are Difficult, but Cuts Also Have Little Room

• Warsh took office in May and was rumored to be hawkish. Around the month before and after his arrival, markets were guided toward rate-hike expectations on the grounds of controlling inflation and preserving Fed independence. Li Xiang added that another Fed governor said that worsening inflation would support a hike. Li Feng called it “still managing expectations.” He said he did not know whether further hikes would continue to raise the Treasury’s long-term interest bill; Li Xiang said they definitely would. Under current inflation, there is little room for cuts either, “unless you are forced to—for example, to rescue the market.” But a rescue would mean capital markets were already falling rapidly.
• The risk-free-rate anchor has moved higher. If 10-year and 30-year yields reach 5%, growth stocks need to offer growth expectations “far above that level, and they have to be sustained” to keep capital. “Otherwise, I may not be bullish anymore. I may have to sell.”

16. Three Black Swans and the Tech Giants Scrambling for Liquidity

• The first black swan is the yen. Once yen liquidity against the US dollar peaks and reverses, it is “just an issuance problem for Treasuries, but a much bigger black swan for other types of risk assets,” especially leveraged risk assets. Li Feng also clarified that Treasuries will not collapse: yields may experience a period of sharp volatility, but Treasuries are local-currency debt, and the US can at least issue money to repay them.
• The second is private debt financing for data centers. America’s most profitable companies are shifting from hidden buyers of Treasuries to issuers competing for liquidity. Google’s cash flow was expected to turn negative only in Q4, but it was already negative in Q2. Most large internet companies, with Microsoft the exception, are now cash-flow negative. Rates in Oracle’s private data-center debt market continue to rise. A blowup could trigger a cascading selloff, like the bonds issued in Hong Kong by Chinese property developers.
• The third is Iran’s control of the Strait of Hormuz and the resulting halt to shipping. The global economy’s ability to absorb shocks is far weaker than it was 2.5 months ago. “If it lasted another, say, 70 days, the commodities market might collapse.” The US believes it cannot regain control of the strait without regime change—“a bit like China’s rare-earth strategy: using a very small card to control a very large game.” Li Feng sees China as relatively safe: since 2017, its strategic oil reserves have risen from 27 days to more than 4 months, supply sources have diversified, and many data centers use green power. Japan and South Korea are less clear.
• The structural warning is that the negative correlation between Treasury yields and US stocks is near a historical extreme. “Black swans never start with earnings; they generally start from outside.” Li Xiang asked whether the market might simply need a reason to fall—“you put out any random piece of news, and it immediately falls for you.” Li Feng’s answer: “That’s exactly how it is.”

17. The Cycle Playbook: The Standard Script for Dollar Assets

• The total-accounting exercise is straightforward: US equities rose from just over $30T in 2019 to roughly $70T, requiring “somewhere around $5T, $6T, $7T, or $8T” of net inflows. US debt rose from $21T to $40T, and every bond issued corresponds to purchasing capital. Global debt is moving from $250T toward more than $350T. The $15T-$20T of base money released by central banks in 2020-2021 was expanded 3-4x through the money multiplier. That is the total stock of cash created.
• The standard script is: rate-hike expectations, followed by an initial decline in risk assets; hikes are delivered and continue; the dollar strengthens and capital flows in. If a crisis that disrupts the world is added—along with a major economy being knocked out, as in the Southeast Asian financial crisis or the European debt crisis—inflows accelerate. Once capital has been attracted to a sufficient degree, the cycle tops out. Rate-cut expectations then produce a short-term rebound, but the moment cuts are delivered or a weaker-dollar trend becomes established is when outflows begin. The strategic lesson for war is that direct US involvement must be brief; if the US does not enter directly, duration matters less, the impact may be broader, and the outcome is more favorable for the dollar.
• The previous major cycle, from the end of 2021 through 2025, had all 3 ingredients, but the country most likely targeted—China—was not knocked out. FX controls, reserves, and capital-account controls helped China survive the difficulties of 2023 and 2024. The operating conclusion for a stock market in the inventory phase is that the odds of claiming unique insight and finding something that rises even as the market plunges are very low. Some speculate that capital pulled out after leverage in Japan and South Korea is broken could produce at most a small rebound in a US equity market worth more than $70T. The Treasury and Fed have introduced arrangements allowing Treasuries to be pledged for dollar liquidity, intended to prevent large-scale Japanese selling. The other side of all this is gold, which also helps explain the recent sharp swings in gold prices.

18. Why China Is Difficult to Forecast in One Direction: Stabilization and Excess Savings

• The stabilization playbook is clear: before early July, Huijin and China Securities Finance had sold more than 90% of their positions, “bringing all the ammunition back.” They then bought broad-market ETFs aggressively whenever the market was oversold and sold the positions back once markets stabilized or even rose intraday. “The only purpose is to prevent excessive moves in either direction.” Li Feng suspects trigger levels and automated execution. Foreign investors remain underweight, or not even at neutral weight, in Hong Kong stocks, while the earlier rally was driven mainly by southbound flows. Chinese assets therefore will not suffer a massive shock simply because capital is suddenly withdrawn, which is why he is reluctant to hedge aggressively with short index futures.
• The variable the world has never seen before is the gradual activation of savings. Estimates of excess savings range from RMB10T-plus to more than RMB30T. China also faces the uncertainty of whether the household savings rate will return to normal rather than remain elevated. If excess savings are released, “wherever that money goes, it will create relatively large and sustained inflows.”

19. Manufacturing Bargaining Chips and a Possible September US Visit

• The recent sequence was that the US sanctioned a group of companies allegedly linked to “forced labor,” followed 2 days ago by reports that Washington would restrict Chinese optical modules. China then retaliated in kind. Li Feng guessed that “the US has always created bargaining chips for itself before negotiations,” possibly in preparation for key talks over the tariff framework and related issues.
• If the timing were set for September and a leaders’ visit to the US went ahead, Li Feng believes it would “definitely be more beneficial than harmful” for Trump’s midterm election prospects. In theory, the US may want the visit more than China does. But China did not treat the chips created by the US as direct negotiating leverage this time and instead retaliated, which may have lowered expectations of a US visit. He stressed that this is only a guess.

20. Primary Markets and Hong Kong Stocks: Narrow Hot Spots and the Break-Below-Issue-Price Test

• The primary market “looks very much like the secondary market at the end of a cycle”: hot spots rotate rapidly, heat builds intensely in a short period, and the concentration is at the intersection of AI and Chinese policy—robots, embodied intelligence, world models, quantum computing, and controllable nuclear fusion. Li Feng’s view is “a little bit certain but also a little bit uncertain”: assuming the capital cycle has ended or a bubble has been pricked and the US market suffers a sharp fall, Chinese investment could spread across the broader “AI+”—AI+consumer, AI+hardware, and AI+industrial manufacturing—toward applications that can actually use AI and make money, rather than the most imaginative themes.
• One of his 4 calls from the start of the year remains unresolved: a Hong Kong dollar regime shift requires a weaker dollar and a broad consensus around that view, while the RMB must not weaken; those conditions have not yet been met. Hong Kong-listed star companies have mostly fallen sharply as lockups approach expiration. His personal, “not guaranteed to work” indicator is the first-day and seven-day break-below-issue-price rate. It used to be just over 10%—roughly 1 of every 8 offerings broke issue price, while the rest were chased, making a 50% gain in a month more attractive than a 5% return over a year. The rate did rise somewhat in July, though it is too early to know whether that will persist. If it does, Hong Kong stocks “may finally become more rational and more investment-oriented.” Another support came when the PBOC governor, attending a Hong Kong event in July for the launch of RMB-related financial derivatives, said FX reserves would be used to support the Hong Kong market; Hong Kong stocks rose for a period afterward.